Stock investors are used to splits, but why all the reverse splits in ETFs?
It’s surprisingly common. ProShares did reverse splits on 26 of its ETFs in the past year. Direxion has done eight. Vanguard reverse split one of its most popular funds, the Vanguard S&P 500 ETF (VOO | A-97).
The most recent reverse split was the 1-for-2 one on the incredibly popular First Trust Short Maturity ETF (FTSM), which coincided with a surprise 10 basis point fee cut.
In normal equity land, reverse splits are often seen as a kind of last-resort window dressing for a failing company. Your stock price is headed toward a dollar, nobody wants it, so you do a 1-for-10 reverse split and—voila!—you’re respectable again and you’ve avoided delisting.
But when VOO reversed, it went from an already-fat $80 price (known as the “handle,” because it’s the part that hangs off the left side of decimal point) to $160. FTSM went from $30 to $60. I get emails every time something like this happens that ask two questions: Why; and, should investors care?
Why Do The Reverse Splits?
Most long-term investors don’t really care much about what the actual dollar price of a share of something is. If I want to put $10,000 to work in a particular ETF or stock, it’s convenient if the share price is small, because it lets me get close to that round dollar amount. If, for instance, I want to buy $10,000 of the SPDR S&P 500 ETF (SPY | A-98), which has a handle of $204, I’m going to end up with either 49 shares and $9,996 of exposure, or 50 shares and $10,200.
At the extreme edges of individual stocks, it becomes hard to actually make a rational investment for some investors: Berkshire Hathaway A shares trade at $219,300 apiece—a higher price than the value of many investors’ entire portfolios. But, for the most part, it’s a marginal difference.
But there are reasons that having a high share price can be beneficial, especially if you’re any kind of active trader or institutional investor.
The first is commissions. While most retail investors trade through discount brokers that charge under $10 a trade, most institutions think by-the-share.
Back when Matt Hougan and I ran mutual funds, our trading desk paid something like 2 cents a share on every trade. In return, we got human beings on the phone to help us work trades, access to both the occasional hot initial public offering allocation and research from our trading partners. While commissions have come down (often less than a penny a share), they’re still a real cost of doing business.
In the ETF space, where hundreds of institutions are often trading the most popular funds with rapid-fire 100-share lots, those commissions can add up quickly. It may only be a buck to trade those $100 shares, but if you’re fancy black box is doing it 10,000 times a day … you get the picture.
The second problem is spreads. While sub-penny trading does happen off-exchange, the vast majority of trading still happens in dollars and cents. Most of the top 100 ETFs ranked by volume trade with as tight a spread as possible—1 penny wide.
That means if the true fair value of something is $100, you’ll pay $100.01 to buy it and maybe can sell it for $100 even—though that spread obviously sloshes around constantly as supply and demand dictates which side of “fair” the market wants to be. But if those shares have a handle of $10 instead of $100, your spread costs you 10 times as much.
Don’t think this matters? Let’s consider a hypothetical example:
A Pithy Hypothetical
Let’s say I want to put $1 million to work in Japan for a week, because I have some hunch (not that I’d recommend that). I look at the market for the most liquid Japan ETFs and I find the iShares MSCI Japan ETF (EWJ | B-97) and the WisdomTree Japan Hedged Equity ETF (DXJ | B-63). Both of these ETFs trade hundreds of millions of dollars a day—my $1 million just isn’t going to move prices around.
DXJ costs 48 basis points a year, or $48 for each $10,000; while EWJ costs 50 basis points. I’m smart enough to know that 2 basis points over a one-week holding period won’t make a darn bit of difference, and will be overwhelmed by the differences in holdings—especially considering that DXJ is currency hedged).
But then I get to the transaction math. As I write this, EWJ has a price of $11.77. DXJ has a price of $54.36. Both trade a penny wide, all day, every day. Let’s imagine I pay half-penny a share in commission. Here’s how that looks to me:
|Number of Shares/$1M||84,962||18,396|
|… as % of trade||0.25%||0.06%|
You can argue with my assumptions all you like—you can argue that you should only be a tenth of a penny a share, or that you can negotiate under a penny on the spreads. Whichever numbers you crunch, the math ends up being the same: DXJ ends up looking a lot less expensive to trade.
There are some other ETF nuances here too: DXJ has a creation-basket size of 50,000 shares, or $2.7 million. EWJ has a 600,000-share creation basket, or $7 million. In a less liquid ETF, those might make these two funds have significantly different premiums and discounts during volatile periods. In this case, both funds trade like water, so it’s less relevant.
Handle Is A Choice
The takeaway here is that ETF issuers get to choose how they want their ETFs to trade:
- You want tight trading to adhere to fair value? Keep your basket size low.
- You want to attract daily volume from institutions? Keep your handle size high.
- You want a lot of allocations from mom-and-pop investors? Maybe the lower handle will net you a bit more marginal flow.
There’s no magically correct answer here. It’s a choice. Luckily, there are so many ETFs covering every corner of the market, chances are you get to make a choice too.
At the time this article was written, the author held no positions in the securities mentioned. You can reach Dave Nadig at [email protected], or on Twitter @DaveNadig.